In situ generated catalyst system to convert biomass-derived levulinic acid to γ-valerolactone

Article abstract of DOI:10.1002/cctc.201500115

This is the first report of HCl/ZrO(OH)₂ catalysts prepared in situ by the autonomous decomposition of ZrOCl₂·8 H₂O in levulinic acid (LA)/2-butanol solution, which catalyzed the esterification of LA in tandem with hydrocyclization to γ-valerolactone (GVL) by Meerwein-Ponndorf-Verley (MPV) reduction without the use of external H₂. A maximum GVL yield of 92.4 % from neat LA and a GVL formation rate of 1092.2 μmol g^-1 min^-1 were achieved in 2-butanol at 240 °C in 2 h. The in situ generated ZrO(OH)₂ was characterized comprehensively and its unexpected catalytic efficiency was attributed mainly to its extremely high surface area. A crude LA stream from the acid hydrolysis of cellulose was extracted into 2-butanol and subjected to this catalyst system to give a GVL yield of 82.0 % even in the presence of humins.

Full text of DOI:10.1002/cctc.201500115

DOI: 10.1002/cctc.201500115
Full Papers
In Situ Generated Catalyst System to Convert Biomass-
Derived Levulinic Acid to g-Valerolactone
Xing Tang,[a] Xianhai Zeng,[a] Zheng Li,[a] Weifeng Li,[a] Yetao Jiang,[a] Lei Hu,[b] Shijie Liu,[c]
Yong Sun,*[a, d] and Lu Lin*[a]
This is the first report of HCl/ZrO(OH)2 catalysts prepared
in situ by the autonomous decomposition of ZrOCl2·8H2O in
levulinic acid (LA)/2-butanol solution, which catalyzed the
esterification of LA in tandem with hydrocyclization to g-val-
erolactone (GVL) by Meerwein–Ponndorf–Verley (MPV) reduc-
tion without the use of external H2. A maximum GVL yield of
1092.2 mmolgꢀ1 minꢀ1 were achieved in 2-butanol at 2408C in
2 h. The in situ generated ZrO(OH)2 was characterized compre-
hensively and its unexpected catalytic efficiency was attributed
mainly to its extremely high surface area. A crude LA stream
from the acid hydrolysis of cellulose was extracted into 2-buta-
nol and subjected to this catalyst system to give a GVL yield of
82.0% even in the presence of humins.
92.4% from neat LA and
a GVL formation rate of
Introduction
Cellulosic biomass, the most abundant available biomass re-
source on earth, is receiving attention with the goal to pro-
duce chemicals and liquid fuels to alleviate the overwhelming
dependence on nonrenewable fossil resources.[1] It is known
that the chemocatalytic conversion of cellulosic biomass to
various platform molecules, such as 5-hydroxymethylfurfural
(HMF),[2] furfural,[3] and g-valerolactone (GVL),[4] promises to be
a key step to realize this fascinating prospect. One such plat-
form compound is GVL that could be synthesized by the hy-
drocyclization of biomass-derived levulinic acid (LA) and its
esters. GVL has received the attention of society and industry
because of its potential applications: 1) GVL can be employed
as a sustainable carbon source for the manufacture of hydro-
carbon transportation fuels;[5] 2) GVL can be converted to vari-
ous polymeric monomers for Nylon production;[6] 3) value-
added chemicals can be synthesized from GVL, such as 2-
adipic acid,[9] and aromatic hydrocarbons;[10] 4) GVL can be
used instead of water as an eco-friendly and renewable solvent
for the production of furfural,[11] LA,[12] and sugars[13] with unex-
pectedly high yields from lignocellulosic biomass. Recently,
GVL was applied as a renewable solvent in organic synthesis,
such as the Hiyama reaction[14] and Sonogashira reaction.[15]
In recent years, multiple catalytic strategies have been devel-
oped to produce LA from biomass materials with high yields in
the lab or on a pilot scale.[16] Therefore, the future role of GVL
in biorefinery will depend strongly on the efficient conversion
of biomass-derived LA to GVL through an affordable and con-
cise process. To date, numerous precious- or base-metal cata-
lysts have been employed to hydrogenate LA to GVL under an
external reducing atmosphere (H2).[4b,17] Among these catalysts,
Ru-based catalysts are superior to other noble-metal catalysts
for the selective hydrogenation of LA to GVL.[6c,18] However,
both precious-metal catalysts and the management of external
molecular H2 are expensive. In this regard, the noble-metal-free
or H2-independent production of GVL from biomass-derived LA
is highly desirable.
methyltetrahydrofuran,[7]
methyl
4-methoxypentanoate,[8]
[a] X. Tang,+ Assist. Prof. X. Zeng,+ Z. Li, W. Li, Y. Jiang, Assist. Prof. Y. Sun,
Prof. L. Lin
College of Energy, Xiamen University
Xiangan South Road, Xiangan District
Xiamen(361102), Fujian Province (P.R. China)
E-mail: sunyong@xmu.edu.cn
Recently, a catalytic transfer hydrogenation (CTH) strategy
for GVL synthesis by Meerwein–Ponndorf–Verley (MPV) reduc-
tion has received increasing attention. Dumesic et al. reported
that alkyl levulinates (ALs) could be subjected to MPV reduc-
tion to yield GVL using ZrO2 and alcohols as the catalyst and
in situ H donors, respectively.[19b] This strategy provides a cost-
effective alternative for GVL production in which precious
metals and H2 are replaced by low-cost catalysts and alcohols,
respectively. However, LA displayed a very poor performance
for MPV reduction using ZrO2 as the catalyst[19b] because organ-
ic acids inhibit MPV reduction to produce GVL.[20] Notably, in-
tractable ammonium hydroxide or alkaline liquor is environ-
mentally unfriendly and high-temperature calcination is
energy-intensive but essential for the preparation of these
metal oxides by conventional precipitation and calcination
lulin@xmu.edu.cn
[b] Dr. L. Hu
Jiangsu Key Laboratory for Biomass-Based Energy and Enzyme Technology
Huaiyin Normal University, Huaian 223300 (P.R. China)
[c] Prof. S. Liu
College of Environmental Science and Forestry
State University of New York
1 Forestry Drive, Syracuse, NY 13210 (USA)
[d] Assist. Prof. Y. Sun
Key Laboratory of Biomass Energy and Materials of Jiangsu Province
Nanjing 210042 (P.R. China)
[+] These authors contributed equally to this work.
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cctc.201500115.
ChemCatChem 0000, 00, 0 – 0
1
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
methods. With regard to these issues, we have developed an
in situ generated catalyst system to convert biomass-derived
LA to GVL, in which HCl/ZrO(OH)2 catalysts were prepared
in situ by the autonomous decomposition of ZrOCl2·8H2O in
LA/2-butanol (2-BuOH) solution and then catalyzed the esterifi-
cation of LA in tandem with the hydrocyclization of AL to GVL
using 2-BuOH as an in situ hydrogen source by MPV reduction.
The catalytic strategy described above is environmentally
friendly and economical because unit operations that include
the ex situ preparation of catalyst and the introduction of ex-
ternal H2 are eliminated.
sorbed a certain amount of organics and had C contents of
0.2–27.1 wt% (Table 1). The atomic ratios of metal (M)/O/Cl of
the recovered catalyst solids were altered dramatically com-
pared with the parent salts. For instance, the atomic ratio of
M/O/Cl was measured as 23.69:72.30:4.01, 20.81:73.88:5.31,
and 23.25:73.16:3.59 if the recovered solid was derived from
AlCl3·6H2O, SnCl2·2H2O, and ZrOCl2·8H2O, respectively (Table 1,
entries 7, 8, and 10). In these cases, the Cl content of the re-
covered solid became negligible compared to that of the
parent salts. Based on these observations, we can infer that
these recovered solid catalysts no longer exist in the form of
the parent salts.
Results and Discussion
Mechanism for the in situ formation of HCl/ZrO(OH)2
catalysts in alcohol
Study of various salts as catalyst precursors
The recovered catalyst derived from ZrOCl2·8H2O is composed
of Zr and O in a ratio (23.25:73.16) that is roughly consistent
with a formula of ZrO(OH)2 (1/3), and the content of Cl was
negligible (Table 1, entry 10). FTIR spectra of these recovered
catalyst solids show absorption peaks centered at n˜ =1540 and
1450 cmꢀ1 (Figure 1, top), which
Among various metal salts used, Zr salts, which include
ZrOCl2·8H2O and Zr(NO3)4·5H2O, showed unparalleled per-
formance as catalyst precursors, and GVL yields as high as 84.5
and 72.3% were obtained at 2408C in 1 h, respectively
(Table 1, entries 10 and 11). No GVL was detected if
are representative of ZrO(OH)2.[20]
Table 1. Catalytic conversion of LA to GVL in 2-BuOH using various salts as the catalyst precursors and elemen-
tal analysis of the recovered catalyst solids.
In the thermogravimetric analy-
sis (TGA) curves (Figure 1,
bottom), the catalyst solid recov-
ered from 2-BuOH shows a total
weight loss of 29.5% during
heat treatment. This value
is similar to 29.4%, which is
the theoretical weight loss of
ZrO(OH)2 (12.8%), carbon depos-
its (14.6%), and water absorbed
physically on the surface of the
recovered catalyst (2%). The X-
ray photoelectron spectra (XPS)
also corroborate that the in situ
generated catalyst in 2-BuOH is
composed mainly of Zr, O, and C
(Figure 2). We can thus conclude
that the recovered catalysts that
originate from ZrOCl2·8H2O exist
in the form of ZrO(OH)2. More-
Entry
Precursor salt
Carbon content[a]
C [wt%] H [wt%]
Atomic ratio
M/O/Cl[b]
XLA
[%]
YSBL
[%]
YGVL
[%]
1
2
3
4
5
6
7
8
blank
–
0.7
11.4
0.2
0.5
27.1
13.5
4.1
–
–
59.3
59.1
98.0
67.8
99.3
93.1
99.9
99.2
99.4
99.9
99.9
80.3
44.8
39.7
64.1
46.4
65.5
48.2
44.3
35.4
59.0
11.7
19.0
74.3
1.0
8.9
BaCl2·2H2O
MgCl2·6H2O
CuCl2·2H2O
MnCl2·2H2O
CrCl3·6H2O
AlCl3·6H2O
SnCl2·2H2O
SnCl4·5H2O
ZrOCl2·8H2O
Zr(NO3)4·5H2O
Zr(SO4)2·4H2O
2.9
5.3
0.9
1.2
4.0
4.3
1.2
1.4
3.5
3.4
2.7
34.88:4.44:60.67
23.05:63.34:13.61
46.95:16.47:36.58
30.40:34.58:35.01
28.02:67.35:4.62
23.69:72.30:4.01
20.81:73.88:5.31
20.56:74.81:4.63
23.25:73.16:3.59
23.05:76.95
11.4
11.7
16.8
20.0
46.1
56.0
25.2
84.5
72.3
0.0
9
4.2
10
11
12
14.6
19.6
3.6
13.59:77.88:8.53[c]
Reaction conditions: LA (43 mmol), salt precursors (5 mol%, relative to LA) and 2-BuOH (95 g) were heated to
2408C for 1 h. [a] Determined by using an Elementar Vario EL III (Germany). [b] Measured by EDS. [c] Atomic
ratio of Zr/O/S.
Zr(SO4)2·4H2O was used, and sec-butyl 4-oxopentanoate (SBL)
was confirmed to be the main product rather than GVL. This is
formed by the esterification reaction between LA and 2-BuOH
(Table 1, entry 12). Apparently, hydrogen transfer was sup-
pressed thoroughly in this case, probably because of catalyst
poisoning in the presence of sulfur, which agrees well with
a previous report.[19a] Moderate GVL yields of 46.1 and 56.0%
were obtained if AlCl3·6H2O and SnCl2·2H2O were used as the
catalyst precursors, respectively (Table 1, entries 7 and 8). In
the blank test and with the use of other salts, SBL was con-
firmed as the major product with GVL yields below 20%
(Table 1, entries 1–6).
over, the liquid product was concentrated by rotary evapora-
tion, and white floc appeared immediately if 0.01m AgNO3 so-
lution was added to the concentrated 2-BuOH solution, which
indicates that there was free Clꢀ in the liquid products. Conse-
quently, we believe that ZrOCl2·8H2O decomposed into
ZrO(OH)2 and HCl in LA/2-BuOH solution under the applied re-
action conditions.
XRD and selected area electron diffraction (SAED) patterns
showed that the in situ generated ZrO(OH)2 existed in an
amorphous state (Figures S1 and S2). Unexpectedly, in situ
generated ZrO(OH)2 in LA/2-BuOH solution had a high specific
surface area of 351 m2 gꢀ1 (Table 2, entry 5) compared with
that of ZrO(OH)2 (191.7 m2 gꢀ1) and ZrO2 (157.2 m2 gꢀ1) pre-
pared by a conventional precipitation method followed by
After reaction, the solid catalysts that originate from the salt
precursors were recovered by filtration. These materials ab-
&
ChemCatChem 0000, 00, 0 – 0
www.chemcatchem.org
2
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Figure 1. FTIR spectra (top) and TGA profiles (bottom) of in situ generated
ZrO(OH)2 catalysts recovered from different alcohols. Reaction conditions: LA
(43 mmol), ZrOCl2·8H2O (5 mol%, relative to LA), and alcohols (95 g) were
heated to 2408C for 1 h under 1 MPa N2.
Figure 2. XPS spectra of the in situ generated ZrO(OH)2 recovered from
2-BuOH (top) and deconvolution of the C1s emission line (bottom). Reaction
conditions are the same as described in Figure 1.
was used as the catalyst under otherwise identical re-
Table 2. The effects of substrates and alcohols (H donors) on the composition and
action conditions.
specific surface area of in situ generated ZrO(OH)2 catalysts.
To gain more insight into the formation of the
ZrO(OH)2 catalyst with
a
high surface area,
Entry
Alcohol
Substrate
Carbon content[a]
Atomic ratio
Zr/O/Cl[b]
SBET
[m2 gꢀ1
]
ZrOCl2·8H2O with different substrates and alcohols
was reacted at 2408C. The in situ generated catalysts
recovered from different alcohols showed a similar IR
absorption profile to that of ZrO(OH)2 (Figure 1, top),
and specific surface areas that ranged from 303 to
363 m2 gꢀ1 were measured that depended on the al-
cohol used (Table 2). However, no solid catalyst was
obtained if the reaction was conducted in methanol
(MeOH). The dielectric constant of a solvent is associ-
ated strongly with the solubility of salts in the sol-
vent and decreases in the order of water (80.37)>
MeOH (33.30)>EtOH (25.09)>2-PrOH (19.52)>
1-BuOH (17.90)>2-BuOH (16.68).[21] Kim et al. found
that the dielectric constant of an aqueous salt solu-
H [wt%]
C [wt%]
1
2
3
4
5
6
7
8
2-BuOH
EtOH
blank
LA
LA
LA
LA
EL
GVL
acetone
3.2
8.8
12.7
8.3
14.6
11.1
9.5
1.6
2.0
3.2
2.7
3.5
2.2
2.1
1.3
–
194
363
321
303
351
204
252
–
21.4:76.8:1.8
19.1:78.0:2.9
26.3:71.3:2.4
23.3:73.2:3.5
–
–
–
2-PrOH
1-BuOH
2-BuOH
2-BuOH
2-BuOH
2-BuOH
2.1
Reaction conditions: substrate (43 mmol), ZrOCl2·8H2O (5 mol%, relative to substrate),
and alcohols (95 g) were heated to 2408C for 1 h under 1 MPa N2. [a] Determined by
using an Elementar Vario EL III (Germany). [b] Measured by EDS.
calcination.[19a,20] The higher specific surface area of the in situ
generated ZrO(OH)2 catalyst could provide more available
active sites that are exposed to substrates to result in a better
catalytic performance for GVL production by MPV reduction. A
recycling test of the in situ generated ZrO(OH)2 was conducted
in 2-BuOH without HCl. An LA conversion of 62.4% and a GVL
yield of 37.0% were achieved at 2408C in 1 h. In contrast, only
25.6% LA was converted to GVL if ex situ prepared ZrO(OH)2
tion decreased greatly with the addition of alcohols, which re-
sulted in the supersaturation of the salt solution and ZrO(OH)2
precipitation.[22] Precipitation also takes place upon heating the
alcohol/aqueous salt solution because the dielectric constant
of the mixed solution is decreased considerably with increasing
temperature.[23] However, MeOH has a high dielectric constant
and cannot be decreased enough to make precipitation occur
even at a high temperature.
ChemCatChem 0000, 00, 0 – 0
www.chemcatchem.org
3
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
Figure 3. SEM images of in situ generated ZrO(OH)2 catalysts recovered from
various alcohols: A) EtOH, B) 2-PrOH, C) 1-BuOH, and D) 2-BuOH. Reaction
conditions are as described in Table 2.
Figure 4. SEM images for the recovered catalyst solids in the presence of dif-
ferent substrates: A) blank test, B) GVL, C) EL, and D) acetone. Reaction con-
ditions are as described in Table 2.
SEM images reveal that the catalyst recovered from EtOH is
composed of ultrafine primary particles (Figure 3). However,
spherical secondary particles with a diameter of 200–500 nm
are observed if ZrO(OH)2 was generated in 2-PrOH or 2-BuOH.
ZrO(OH)2 recovered from 1-BuOH shows large, clumpy aggre-
gations and the lowest surface area (303 m2 gꢀ1) compared
with that generated in other alcohols (Table 2).
generated ZrO(OH)2 catalysts, that is, the in situ formed
ZrO(OH)2 catalyst with a high specific surface area has a high C
content if LA was applied as the substrate. These carbon de-
posits on the surface of the recovered ZrO(OH)2 catalysts could
be attributed mainly to the adsorption of the intermediate
(SBL) and/or GVL because the characteristic IR absorption peak
of a carbonyl group (n˜ =1710 cmꢀ1; Figure 1) and the binding
energy of a C=O bond (Figure 2) were observed.
However, the specific surface area and morphology of the
in situ generated catalysts showed a strong dependence on
the substrates. In a control test, the specific surface area of the
recovered catalyst was measured to be 194 m2 gꢀ1 in the ab-
sence of substrate (Table 2, entry 1). The surface area was in-
creased significantly to 351 m2 gꢀ1 in the presence of LA in 2-
BuOH (Table 2, entry 5) and to 204 and 252 m2 gꢀ1 in the pres-
ence of EL and GVL, respectively (Table 2, entries 6 and 7). SEM
images show that catalyst solids aggregate to much larger and
rodlike particles if the catalyst was generated in the absence of
LA in 2-BuOH (Figure 4), in contrast, smaller spherical particles
are observed in the presence of LA (Figure 3). According to
Derjaguin–Landau–Verwey–Overbeek (DLVO) theory, the mor-
phology of colloids is dominated mainly by electrostatic repul-
sion forces among particles.[22] The mechanism of MPV reduc-
tion was elucidated by considering that the adsorption of sub-
strate and H donors onto the surface of solid catalyst took
place before hydrogen transfer.[20] At this point, the absorbed
LA and H donors probably enhanced the electrostatic repul-
sion forces between the in situ formed ZrO(OH)2 particles and
then prevented further agglomeration in 2-BuOH.
In short, the dispersity and morphology of in situ generated
ZrO(OH)2 particles relied strongly on the solvent and substrate
used. The solvents and substrates also had a substantial influ-
ence on the GVL yield, which will be discussed at the next sec-
tion.
Reaction pathway of the conversion of LA to GVL
In the recycling test, we found that the recovered ZrO(OH)2
catalyst showed a low catalytic activity compared to its parent
salt; only a GVL yield of 37.0% and an LA conversion of 62.4%
were achieved with the recovered material. However, the GVL
yield was increased sharply to 71.6% with the quantitative
conversion of LA if a small amount of HCl (1.5 mmol) was
added under the same reaction conditions. LA could not be
converted immediately to 4-hydroxyvaleric acid (HVA) by MPV
reduction under the reaction conditions discussed here, proba-
bly because MPV reduction is restrained by organic acids.[24]
Apparently, in situ generated HCl along with ZrO(OH)2 that
comes from the decomposition of ZrOCl2·8H2O could eliminate
the negative impact of the organic acid by the esterification of
LA with 2-BuOH. SBL was one of the main compounds in the
liquid product. This observation indicates that LA first under-
goes esterification to SBL, which is followed by the CTH of SBL
to sec-butyl 4-hydroxypentanoate (SHPB) and the lactonization
of SHPB to yield GVL in 2-BuOH (Scheme 1). No intermediate
SHPB was observed in the products because the lactonization
of SHPB to GVL is fast at a temperature above 1508C in the
In addition, a relatively higher C content in the recovered
ZrO(OH)2 catalysts was recorded if LA was used as the sub-
strate. The C content of the recovered ZrO(OH)2 catalyst was
measured as 14.6 wt% if LA was added (Table 2, entry 5),
whereas the C content decreased sharply to 3.2 or 2.1 wt% in
the blank test or if acetone was used, respectively (Table 2, en-
tries 1 and 8). This trend is in good agreement with the sub-
strate dependence of the specific surface area of the in situ
&
ChemCatChem 0000, 00, 0 – 0
www.chemcatchem.org
4
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
1 h with a catalyst loading of 5 mol% (Table 3, entries 9 and
16). Under the same reaction conditions, GVL yields were de-
creased to 76.4 and 64.2% if EtOH and 1-BuOH were applied
as the solvent and H donor, respectively (Table 3, entries 5 and
12). However, the GVL yield in 2-BuOH was much lower than
that in 2-PrOH at a relatively low reaction temperature. For ex-
ample, a GVL yield of 62.0% was achieved in 2-PrOH at 2008C
in 1 h with 3 mol% ZrOCl2·8H2O, in contrast, only 27.5% of LA
was converted to GVL in 2-BuOH under identical reaction con-
ditions (Table 3, entries 6 and 13). This result can be ascribed
mainly to the stronger steric effect of 2-BuOH than that of 2-
PrOH, and a high temperature was beneficial to overcome the
steric effect of 2-BuOH. If the reaction temperature was in-
creased to 2408C, a GVL yield of 84.5% was obtained in 2-
BuOH, which is similar to that obtained in 2-PrOH under the
same reaction conditions (Table 3, entries 9 and 16). In addi-
tion, an elevated temperature improved the GVL formation
rate markedly from 822.7 to 1806.8 mmolgꢀ1 minꢀ1 if EtOH was
used as the solvent and H donor (Table 3, entries 3–5). A simi-
lar trend was observed if other alcohols were used as the H
donor.
Scheme 1. Proposed reaction pathway for the conversion of LA to GVL by
MPV reduction.
presence of acid catalysts.[25] Therefore, the CTH of SBL to
SHPB was the rate-limiting step for the conversion of LA to
GVL. Moreover, a trace amount of sec-butyl 4-sec-butoxypenta-
noate (SBOP), which was formed by the etherification reaction
between SHPB and 2-BuOH, was detected. If GVL (43 mmol)
and ZrOCl2·8H2O (5 mol%, relative to GVL) were reacted at
2408C for 3 h, a recovery rate of GVL of 93.1% was obtained
and SBOP was detected by GC–MS, which indicates that the
cyclization of SHPB to GVL was reversible and the reaction
equilibrium was preferred significantly to the formation of GVL
under the given reaction conditions.
The CTH of LA to GVL in 2-BuOH was investigated in depth
using in situ generated HCl/ZrO(OH)2 catalysts (Table 4). The
blank experiment gave only an insignificant GVL yield with an
SBL yield of 28.2% at 2008C in the absence of catalyst (Table 4,
entry 1). The yields of GVL and SBL increased to 16.0 and
60.1% if 1 mol% of ZrOCl2·8H2O was added (Table 4, entry 2).
A further increase of the temperature to 2608C and the cata-
lyst loading to 5 mol%, improved the GVL yield gradually from
16.0 to 89.4% to the detriment of the SBL yield, which de-
creased from 60.1 to 3.1% (Table 4, entries 2–5 and 10). How-
ever, if the catalyst loading was further increased to 10 or
15 mol%, the GVL yield decreased to 73.7 or 72.2% with
a slight increase of the SBL yield (Table 4, entries 11 and 12). A
relatively higher concentration of HCl, which origi-
Investigation of the CTH of LA to GVL under various
reaction conditions
MPV reduction is associated strongly with the reducing capaci-
ty of alcohols (H donors), which decrease in the order of
MeOH<EtOH<1-BuOH<2-BuOHꢁ2-PrOH (Table S1).[26] In
MeOH, methyl levulinate (ML) was verified as the dominant
product with a negligible GVL yield (Table 3, entry 1). 2-PrOH
and 2-BuOH were confirmed to be active H donors, which
bring about similar GVL yields of 83.1 and 84.5% at 2408C in
nates from the decomposition of ZrOCl2·8H2O, could
partially suppress the CTH of SBL to GVL (discussed
Table 3. The effect of alcohols (the H donors) on the CTH of LA to GVL.
in the next section). An analogous tendency was also
[b]
[c]
Entry
Alcohol
T
XLA
YLE
YGVL
RGVL
discovered if AlCl3·6H2O or SnCl2·2H2O was employed
[8C]
[%]
[%]
[%]
[mmol gꢀ1 minꢀ1
]
as the catalyst precursor (Table 4, entries 13–18). In
this scenario, the GVL formation rate increased and
peaked at 3528.4 mmolgꢀ1 minꢀ1 (Table 4, entry 5)
and then decreased to 569.0 mmolgꢀ1 minꢀ1 (Table 4,
entry 12).
1[a]
2[a]
3
4
5
6[a]
7
MeOH
EtOH
EtOH
EtOH
EtOH
2-PrOH
2-PrOH
2-PrOH
2-PrOH
1-BuOH
1-BuOH
1-BuOH
2-BuOH
2-BuOH
2-BuOH
2-BuOH
200
200
200
220
240
200
200
220
240
200
220
240
200
200
220
240
99.9
99.9
99.9
99.9
99.9
99.9
99.9
99.9
99.9
99.5
96.5
97.0
67.9
99.9
99.9
99.9
97.6
66.2
40.5
31.6
14.6
32.9
27.5
19.8
9.5
60.7
29.7
12.6
29.0
32.1
24.5
11.7
1.3
28.9
34.8
55.0
76.4
62.0
62.7
68.2
83.1
27.9
51.7
64.2
27.5
56.1
67.8
84.5
–
1140.6
822.7
1300.9
1806.8
2444.5
1482.3
1613.2
1964.9
1099.0
1222.2
1517.7
1084.5
1326.2
1603.5
1996.8
Moreover, the GVL yield increased significantly and
then peaked at 92.4% with a prolonged reaction
time from 0.5 to 2 h at 2408C (Table 4, entries 6 and
7). In this case, the GVL formation rate was
1092.2 mmolgꢀ1 minꢀ1, which is increased by nearly
100 times compared with a previous report in which
Dumesic et al. demonstrated that a GVL formation
rate of only 12.6 mmolgꢀ1 minꢀ1 and a GVL yield of
71% were obtained if LA and 2-BuOH reacted at
2208C over ZrO2 for 16 h.[19b] If the reaction time was
further prolonged to 4 h, the GVL yield decreased
slightly to 86.5% (Table 4, entry 9). Meanwhile,
a slight increase of the yield of the byproduct SBOP
8
9
10[a]
11
12
13[a]
14
15
16
Reaction conditions: 43 mmol LA, 95 g alcohols, catalyst loading (ZrOCl2·8H2O, relative
to LA) 5 mol%, 1 h, 1 MPa N2, 600 rpm. [a] Catalyst loading 3 mol%. [b] YLE is yield of
levulinates, for example, YLE is the yield of methyl levulinate in entry 1. [c] Based on
the mass of ZrO(OH)2.
ChemCatChem 0000, 00, 0 – 0
www.chemcatchem.org
5
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
31.6%, respectively (Table 5, entries 4 and 8). In con-
trast, a relatively low concentration of water (5 wt%)
or HCl (1.5 mmol) impacted the LA conversion slight-
ly and led to an acceptable decrease of GVL yield to
80.6 and 78.5%, respectively (Table 5, entries 2 and
7). A GVL yield of only 60.6% was obtained in the
presence of FA (43 mmol) at 2408C in 1 h, whereas
the GVL yield was improved to 92.9% with a pro-
longed reaction time of 4 h (Table 5, entries 6 and 7).
The negative impact of FA (organic acid) on MPV re-
duction was eliminated by the esterification of FA
with a prolonged reaction time, similar to that of LA.
However, the GVL yield was decreased dramatically
to 8.3% if H2SO4 was introduced into the reactor, and
the consumption of 2-BuOH by dehydration to
butene was observed (Table 5, entry 9). H2SO4 could
promote the dehydration of 2-BuOH and hinder hy-
drogen transfer from 2-BuOH to SBL.
Table 4. The effects of catalyst loading and temperature on the CTH of LA to GVL in
2-BuOH.
[d]
Entry
Catalyst[a]
[mol%]
T
[8C]
t
[h]
XLA
[%]
YSBL
[%]
YGVL
[%]
RGVL
[mmol gꢀ1 minꢀ1
]
1
2
3
4
5
6
7
8
blank
1
200
200
200
220
240
240
240
240
240
260
240
240
240
240
240
240
240
240
1
1
1
1
1
0.5
2
3
4
1
1
1
1
1
1
1
1
1
37.4
84.5
85.0
99.9
99.9
99.9
99.9
99.9
99.9
99.9
99.9
99.9
99.7
99.9
99.3
99.2
99.2
99.9
28.2
60.1
47.5
43.2
32.8
27.9
4.9
4.3
3.4
3.1
15.3
15.4
55.3
44.3
53.7
39.0
35.4
38.8
0.6
16.0
26.5
48.7
59.7
67.2
92.4
92.3
86.5
89.4
73.7
72.2
31.5
46.1
31.1
52.0
56.0
52.2
–
1891.3
1566.2
2878.3
3528.4
3175.5
1092.2
727.7
511.4
2114.5
871.2
569.0
–
2
2
2
5
5
5
5
5
10
15
3
5
10
3
9
10
11
12
13[b]
14[b]
15[b]
16[c]
17[c]
18[c]
–
–
–
–
5
10
There was a significant decrease in the C content
of the recycled catalyst formed in the presence of im-
purities compared with that formed in the presence
of LA only (14.6 wt%; Table 2, entry 5). For example,
the C content of the recovered catalyst decreased to
5.3 or 5.7 wt% if 10 wt% water or 15 mmol HCl was
added to the reactor (Table 5, entries 4 and 8). This
–
Reaction conditions: 43 mmol LA, 95 g 2-BuOH, 1 MPa N2, 600 rpm. [a] ZrOCl2·8H2O
(relative to LA). [b] Using AlCl3·6H2O as the catalyst precursor. [c] Using SnCl2·2H2O as
the catalyst precursor. [d] Based on the mass of ZrO(OH)2.
was observed, which indicates that the decrease of the GVL
yield can be ascribed mainly to the ring-opening of GVL to
SHPB (Scheme 1). 2-sec-Butoxybutane and 2,2-di-sec-butoxybu-
tane were also detected, which were formed by the etherifica-
tion of 2-BuOH and the condensation between butanone with
2-BuOH, respectively. However, only a small amount of solvent
(2-BuOH) was consumed by these side reactions even after re-
action at 2408C for 4 h (Figure S4).
trend is in good agreement with the decrease of the specific
surface area of these recovered catalysts, which implies that
H2O and HCl probably disturb the interaction between LA and
the in situ generated catalysts.
To minimize the negative effect of these impurities, the pre-
treatment of the hydrolysate that comes from the acid-cata-
lyzed hydrolysis of biomass carbohydrates is a prerequisite to
guarantee a high GVL yield. In this study, partial neutralization
and extraction by 2-BuOH were employed to remove the min-
eral acid catalyst and water. The separation of neat LA from
the synthetic hydrolysate was tested by liquid–liquid extraction
systems that comprised an aqueous LA solution and the same
volume of 2-BuOH. The detailed procedure for the extraction
experiments is described in the Supporting Information. The
CTH of crude LA derived from cellulose to GVL in 2-BuOH
With regard to the outstanding catalytic efficiency of in situ
generated HCl/ZrO(OH)2 catalysts, it is promising to utilize this
catalytic strategy for the CTH of crude LA, instead of off-the-
shelf pure LA, which was produced by the acid-cata-
lyzed hydrolysis of biomass carbohydrates.
In general, liquid products that result from the
acid-catalyzed hydrolysis of biomass carbohydrates
are composed mainly of LA (desired product), formic
acid (FA, byproduct), mineral acids (catalysts), H2O
(solvent), humins, and other undetectable intermedi-
ates. In light of the complexity of this hydrolysate,
the conversion of LA to GVL was simulated prelimi-
narily in 2-BuOH in the presence of typical impurities
that are usually present in the hydrolysate of carbo-
hydrates (Table 5).
Table 5. The effects of impurities in 2-BuOH on the formation of the ZrO(OH)2 catalyst
and GVL yield.
Entry Impurity
t [h]
Carbon content[a]
SBET
[m2 gꢀ1
XLA
[%]
YSBL
[%]
YGVL
[%]
]
C [wt%] H [wt%]
1
2
3
4
5
6
7
8
9
5 wt% H2O[b]
1
2
1
3
1
4
1
1
1
–
–
–
–
299
146
319
–
–
98
–
99.9 30.8 68.6
99.9 18.2 80.6
92.0 39.8 51.7
99.4 30.0 68.3
99.9 25.3 60.6
5 wt% H2O[b]
10 wt% H2O[b]
10 wt% H2O[b]
43 mmol FA
10.5
11.5
5.3
12.4
–
12.1
5.7
1.2
3.2
3.0
1.1
3.1
–
3.6
2.7
2.4
43 mmol FA
99.9
5.2 92.9
A relatively high concentration of water (10 wt%)
or HCl (15 mmol) was found to facilitate the aggrega-
tion of ZrO(OH)2 to give rise to a considerable de-
crease of the surface area of the recovered catalysts
to 146 or 98 m2 gꢀ1, respectively. Consequently, the
GVL yields were also reduced sharply to 68.3 and
1.5 mmol HCl
15 mmol HCl
1.5 mmol H2SO4
99.9 18.6 78.5
99.9 58.9 31.6
89.5 57.3
8.3
Reaction conditions: LA (43 mmol), ZrOCl2·8H2O (5 mol%, relative to LA), and 2-BuOH
(95 g) were heated to 2408C for the prescribed duration under 1 MPa N2. [a] Deter-
mined by using an Elementar Vario EL III (Germany). [b] Relative to solvent.
&
ChemCatChem 0000, 00, 0 – 0
www.chemcatchem.org
6
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
addition of NaCl into the aqueous phase promoted the extrac-
tion behavior of 2-BuOH greatly and decreased the solubility
of H2O in the 2-BuOH extract (Table S2), which generates a fa-
vorable environment for the CTH of LA to GVL.
Experimental Section
Chemicals and materials
Levulinic acid (LA, 98%), ethyl levulinate (EL, 98%), g-valerolactone
(GVL, 98%), microcrystalline cellulose (90 mm), and zirconium oxy-
chloride (ZrOCl2·8H2O, 99%) were purchased from Aladdin Reagent
Co. Ltd. (Shanghai, China). All other chemicals were supplied by
Sinopharm Chemical Reagent Co. Ltd. (Shanghai, China) and used
without further purification.
Herein, microcrystalline cellulose was selected as the feed-
stock to produce LA, and an LA yield of 47.8% was achieved.
The detailed procedure for LA production from cellulose and
subsequent extraction by 2-BuOH is described in the Support-
ing Information. Finally, 90.2% of LA and 15.1% of FA were ex-
tracted by 2-BuOH from the hydrolysate of cellulose. A yellow-
brown solution was obtained because humins also dissolved in
2-BuOH in the extraction process, and the mass concentration
of LA in the 2-BuOH extract was approximately 5 wt% (Fig-
ure S6). The resulting 2-BuOH extract and ZrOCl2·8H2O
(5 mol%, relative to LA) were reacted at 2408C, and a GVL
yield of 82.0% (based on LA) was still achieved even in the
presence of humins (Scheme 2). This is the first report of the
production of GVL in a high yield from crude LA by MPV re-
duction using alcohol as the H donor.
Catalytic reactions and sample analysis
All experiments were performed in a 400 mL Hastelloy-C high-pres-
sure reactor (Dalian-controlled Plant, Dalian, China). The reactor
was heated in an adjustable electric stove. The temperature of the
reactor contents was monitored by using a thermocouple connect-
ed to the reactor. Typically, substrate (LA, 5 g), solvent (2-BuOH,
95 g), and catalyst salt precursor (ZrOCl2·8H2O, 5 mol%, relative to
substrate) were charged into the reactor, which was sealed, filled
with N2 (1 MPa), and heated to the prescribed temperature for the
desired reaction time with stirring at 600 rpm. After the reaction,
the reactor was cooled to RT. The solid catalysts (white powder)
were recovered by filtration under vacuum and dried in a vacuum
oven at 608C for 4 h, and the liquid products were centrifuged at
8000 rpm for 5 min. The liquid products were analyzed by GC–MS
and GC, respectively. The conversion and product selectivity were
determined using calibration curves obtained by analyzing stan-
dard solutions. Typically, the GC analysis of each sample was con-
ducted by using an Agilent 7890 series system equipped with
a DB-WAXetr column (30.0 mꢁ0.25 mmꢁ0.25 mm) and a flame ion-
ization detector (FID) that was operated at 2708C. The carrier gas
was N2 with a flow rate of 1.0 mLminꢀ1. The following temperature
program was used: 408C (4 min) and 158Cminꢀ1 to 2508C (5 min).
The qualitative analysis of the products after reaction was conduct-
ed by using a Shimadzu QP2010SE instrument with an Rtx-5 MS
column (30.0 mꢁ0.25 mmꢁ0.25 mm) and EI-MS. The operating pa-
rameters of the GC–MS analysis were in line with that of GC analy-
sis. LA conversion (XLA [%]), GVL yield (YGVL [%]), and GVL formation
rate (RGVL [mmolgꢀ1 minꢀ1]) were calculated according to Equa-
tions (1)–(3):
Scheme 2. Schematic representation of GVL production from cellulose
through the integration of acid-catalyzed hydrolysis, extraction, and subse-
quent CTH process catalyzed by in situ generated ZrO(OH)2 and HCl.
Conclusions
We present an in situ generated catalyst system for the conver-
sion of biomass-derived levulinic acid (LA) to g-valerolactone
(GVL). The mechanism of the formation of the HCl/ZrO(OH)2
catalysts from ZrOCl2·8H2O in alcohols was elaborated, and the
catalytic transfer hydrogenation of LA to GVL catalyzed by this
in situ generated HCl/ZrO(OH)2 catalysts was investigated com-
prehensively using 2-butanol as the solvent and H donor. The
delicate strategy reported here might minimize the costs for
the production of GVL from carbohydrates by avoiding an
energy-intensive LA purification process and the use of noble-
metal catalysts and molecular H2. Especially, tedious and envi-
ronmentally hazardous processes for the preparation of cata-
lysts were eliminated through the spontaneous decomposition
of the inexpensive salt precursors in alcohols. The in situ gen-
erated HCl/ZrO(OH)2 catalysts showed a high catalytic activity
for the catalytic transfer hydrogenation of LA to GVL and satis-
factory robustness against intractable humins that come from
the acid-catalyzed hydrolysis of carbohydrates. In addition, the
in situ generated catalyst system described here is also applica-
ble to the hydrogenation of other biomass-derived molecules,
which include 5-hydroxymethylfurfural and furfural.
Mole of LA in the products
ð1Þ
ð2Þ
ð3Þ
XLA ¼ ð1ꢀ
Þ ꢂ 100
Initial mole of LA
Mole of GVL in the products
ꢂ 100
Initial mole of LA
YGVL
¼
¼
Mole of GVL in the products
RGVL
Mass of ZrOðOHÞ2 ꢂ Reaction time
The detailed procedure for LA production from cellulose and sub-
sequent extraction by 2-BuOH is described in the Supporting Infor-
mation.
Catalyst characterization
XRD patterns were obtained by using a Panalytical X’pert Pro dif-
fractometer using a CuKa radiation source with the following pa-
rameters: 40 kV, 30 mA, 2q=20–908 at a scanning speed of
78 minꢀ1. FTIR spectra were recorded by using a Nicolet 380 spec-
trometer. The morphology of the recovered catalyst powder was
observed by SEM (Hitachi S-4800), equipped with an energy disper-
sive X-ray spectrometer (EDS) that was used to analyze the ele-
ChemCatChem 0000, 00, 0 – 0
www.chemcatchem.org
7
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
Zhang, R. Kumar, A. Foster, S. M. Sen, C. T. Maravelias, R. Malina, S. R. H.
Barrett, R. Lobo, C. E. Wyman, J. A. Dumesic, G. W. Huber, Energy Envi-
ron. Sci. 2014, 7, 1500–1523; d) J. C. Serrano-Ruiz, D. J. Braden, R. M.
West, J. A. Dumesic, Appl. Catal. B 2010, 100, 184–189.
[6] a) J.-P. Lange, J. Z. Vestering, R. J. Haan, Chem. Commun. 2007, 3488–
3490; b) S. Raoufmoghaddam, M. T. Rood, F. K. Buijze, E. Drent, E. Bouw-
man, ChemSusChem 2014, 7, 1984–1990; c) L. E. Manzer, Appl. Catal. A
2004, 272, 249–256.
[7] M. G. Al-Shaal, A. Dzierbinski, R. Palkovits, Green Chem. 2014, 16, 1358–
1364.
[8] Z. Li, Y. Jiang, X. Tang, M. Zuo, X. Zeng, Y. Sun, L. Lin, RSC Adv. 2015, 5,
8297–8300.
mental content (Zr/O/Cl) of the recovered catalyst solid. The con-
tent of each element was calculated by the average value of at
least five scans on the different area of the catalyst. TEM images
and the corresponding SAED patterns were obtained by using
a JEOL JEM-2100 instrument operated at an accelerating voltage of
200 kV. XPS spectra were measured by using a PHI Quantum-2000
electron spectrometer (Ulvac-Phi, Japan) with 150 W monochrom-
atized AlKa radiation (1486.6 eV). TGA was performed by using
a SDT Q600 thermal analyzer under a dynamic N2 atmosphere
(100 mLminꢀ1) in the temperature range of 20–9008C with a heat-
ing rate of 208Cminꢀ1. The surface area, pore volume, and pore
size of the recovered catalysts were calculated by N2 adsorption–
desorption isotherms measured by using a TriStar 3000 with the
BET and Barrett–Joyner–Halenda (BJH) methods. The samples were
degassed at 373 K for 3 h under vacuum before N2 adsorption. Ele-
mental analyses (C and H) were performed by using an Elementar
Vario EL III (Elementar Analysensysteme GmbH, Germany), and
each sample was measured twice to determine the content of C
and H.
[9] S. Van de Vyver, Y. Romꢂn-Leshkov, Catal. Sci. Technol. 2013, 3, 1465–
1479.
[10] Y. Zhao, Y. Fu, Q.-X. Guo, Bioresour. Technol. 2012, 114, 740–744.
[11] a) E. I. Gꢃrbꢃz, J. M. Gallo, D. M. Alonso, S. G. Wettstein, W. Y. Lim, J. A.
Dumesic, Angew. Chem. Int. Ed. 2013, 52, 1270–1274; Angew. Chem.
2013, 125, 1308–1312; b) L. Zhang, H. Yu, P. Wang, Y. Li, Bioresourc.
Technol. C 2014, 151, 355–360.
[12] a) L. Qi, Y. F. Mui, S. W. Lo, M. Y. Lui, G. R. Akien, I. T. Horvꢂth, ACS Catal.
2014, 4, 1470–1477; b) S. Wettstein, D. Martin Alonso, Y. Chong, J. Du-
mesic, Energy Environ. Sci. 2012, 5, 8199–8203; c) D. M. Alonso, S. G.
Wettstein, M. A. Mellmer, E. I. Gurbuz, J. A. Dumesic, Energy Environ. Sci.
2013, 6, 76–80.
[13] J. S. Luterbacher, J. M. Rand, D. M. Alonso, J. Han, J. T. Youngquist, C. T.
Maravelias, B. F. Pfleger, J. A. Dumesic, Science 2014, 343, 277–280.
[14] E. Ismalaj, G. Strappaveccia, E. Ballerini, F. Elisei, O. Piermatti, D. Gelman,
L. Vaccaro, ACS Sustainable Chem. Eng. 2014, 2, 2461–2464.
[15] G. Strappaveccia, L. Luciani, E. Bartollini, A. Marrocchi, F. Pizzo, L. Vac-
caro, Green Chem. 2015, in press.
Acknowledgements
This work was supported financially by the Key Research Program
from Science and Technology Bureau of Xiamen City in China
(3502Z20131016), the National Natural Science Foundation of
China (21106121), the Jiangsu Province Key Laboratory of Bio-
mass Energy and Materials (JSBEM201403), the Key Program for
Cooperation Between Universities and Enterprises in Fujian Prov-
ince (2013N5011), and the Fundamental Research Funds for the
Xiamen University (201312G009). We also appreciate supports
from the China Scholarship Council.
[16] D. W. Rackemann, W. O. S. Doherty, Biofuels, Bioprod. Biorefin. 2011, 5,
198–214.
[17] W. R. Wright, R. Palkovits, ChemSusChem 2012, 5, 1657–1667.
[18] a) Z. Yan, L. Lin, S. Liu, Energy Fuels 2009, 23, 3853–3858; b) J. Tan, J.
Cui, T. Deng, X. Cui, G. Ding, Y. Zhu, Y. Li, ChemCatChem 2015, 7, 508–
512; c) A. Dutta Chowdhury, R. Jackstell, M. Beller, ChemCatChem 2014,
6, 3360–3365.
[19] a) X. Tang, L. Hu, Y. Sun, G. Zhao, W. Hao, L. Lin, RSC Adv. 2013, 3,
10277–10284; b) M. Chia, J. A. Dumesic, Chem. Commun. 2011, 47,
12233–12235; c) X. Tang, Y. Sun, X. Zeng, W. Hao, L. Lin, S. Liu, Energy
Fuels 2014, 28, 4251–4255.
Keywords: biomass · cyclization · hydrogenation · reduction ·
zirconium
[20] X. Tang, H. Chen, L. Hu, W. Hao, Y. Sun, X. Zeng, L. Lin, S. Liu, Appl.
Catal. B 2014, 147, 827–834.
[21] N. Koizumi, T. Hanai, Bull. Inst. Chem. Res. Kyoto Univ. 1955, 33, 14–20.
[22] Y. T. Moon, H. K. Park, D. K. Kim, C. H. Kim, I.-S. Seog, J. Am. Ceram. Soc.
1995, 78, 2690–2694.
[23] G. Akerlof, J. Am. Chem. Soc. 1932, 54, 4125–4139.
[24] C. F. de Graauw, J. A. Peters, H. van Bekkum, J. Huskens, Synthesis 1994,
10, 1007–1017.
[25] H. Y. Luo, D. F. Consoli, W. R. Gunther, Y. Romꢂn-Leshkov, J. Catal. 2014,
320, 198–207.
[1] a) A. Corma, S. Iborra, A. Velty, Chem. Rev. 2007, 107, 2411–2502; b) S.
Liu, J. Bioprocess Eng. Biorefinery 2012, 1, 129–139.
[2] a) L. Hu, G. Zhao, W. Hao, X. Tang, Y. Sun, L. Lin, S. Liu, RSC Adv. 2012, 2,
11184–11206; b) H. Zhao, J. E. Holladay, H. Brown, Z. C. Zhang, Science
2007, 316, 1597–1600.
[3] a) J. P. Lange, E. van der Heide, J. van Buijtenen, R. Price, ChemSusChem
2012, 5, 150–166; b) K. Yan, G. Wu, T. Lafleur, C. Jarvis, Renewable Sus-
tainable Energy Rev. 2014, 38, 663–676.
[4] a) I. T. Horvꢂth, H. Mehdi, V. Fꢂbos, L. Boda, L. T. Mika, Green Chem.
2008, 10, 238–242; b) X. Tang, X. Zeng, Z. Li, L. Hu, Y. Sun, S. Liu, T. Lei,
L. Lin, Renewable Sustainable Energy Rev. 2014, 40, 608–620.
[5] a) J. Q. Bond, D. M. Alonso, D. Wang, R. M. West, J. A. Dumesic, Science
2010, 327, 1110–1114; b) J. P. Lange, R. Price, P. M. Ayoub, J. Louis, L.
Petrus, L. Clarke, H. Gosselink, Angew. Chem. Int. Ed. 2010, 49, 4479–
4483; Angew. Chem. 2010, 122, 4581–4585; c) J. Q. Bond, A. A. Upad-
hye, H. Olcay, G. A. Tompsett, J. Jae, R. Xing, D. M. Alonso, D. Wang, T.
[26] J. C. van der Waal, P. J. Kunkeler, K. Tan, H. van Bekkum, J. Catal. 1998,
173, 74–83.
Received: February 6, 2015
Published online on && &&, 0000
&
ChemCatChem 0000, 00, 0 – 0
www.chemcatchem.org
8
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

FULL PAPERS
X. Tang, X. Zeng, Z. Li, W. Li, Y. Jiang,
L. Hu, S. Liu, Y. Sun,* L. Lin*
&& – &&
In Situ Generated Catalyst System to
Convert Biomass-Derived Levulinic
Acid to g-Valerolactone
In situ generated catalyst system:
HCl/ZrO(OH)2 catalysts are prepared in
situ by the autonomous decomposition
of ZrOCl2·8H2O in levulinic acid
tion of LA was catalyzed in tandem with
hydrocyclization to g-valerolactone
(GVL) by Meerwein–Ponndorf–Verley
reduction.
(LA)/2-butanol solution. The esterifica-
ChemCatChem 0000, 00, 0 – 0
www.chemcatchem.org
9
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ